In order to truly obtain the feature extraction of vibration signals under the strong background noise, the analysis and improvement of empirical mode decomposition (EMD) is carried on. After that, the improved EMD ...In order to truly obtain the feature extraction of vibration signals under the strong background noise, the analysis and improvement of empirical mode decomposition (EMD) is carried on. After that, the improved EMD is applied to the feature extraction of vehicle vibration signals. First, the multi-autocorrelation method is adopted in each input signal,so the noise is reduced effectively. Then, EMD is used to deal with these signals,and the intrinsic mode functions (IMFs) are obtained. Finally, for obtaining the feature information of these signals, the Hilbert transformation and the spectrum analysis are performed in some IMFs. Theoretical analysis and ex- periment verify the effectiveness of the method, which are valuable reference for the same engineering problems.展开更多
To improve the automation level and operation quality of China's beet harvester and reduce the loss due to damaged and missed excavation,this study used a self-developed sugar beet combine harvester and field simu...To improve the automation level and operation quality of China's beet harvester and reduce the loss due to damaged and missed excavation,this study used a self-developed sugar beet combine harvester and field simulation experiment platform,based on the single-factor bench test of the automatic row following system in the early stage,taking hydraulic flow A,spring preload B,and forward speed C which have significant influence on performance indices as test factors,and taking the missed excavation rate,breakage rate and reaction time as performance indices,the orthogonal experimental study on the parameter optimization of the three-factor and three-level automatic row following system with the first-order interaction of various factors was carried out.The results of the orthogonal experiments were analyzed using range analysis and variance analysis.The results showed that there were differences in the influence degree,factor priority order and first-order interaction,and the optimal parameter combination on each performance index.A weighted comprehensive scoring method was used to optimize and analyze each index.The optimal parameter combination of the overall operating performance of the automatic row following system was A 2B 2C 1,that is,the hydraulic flow was 25 L/min,the forward speed was 0.8 m/s,and the spring preload was 198 N.Under this combination,the response time was 0.496 s,the missed excavation rate was 2.35%,the breakage rate was 3.65%,and the operation quality was relatively good,which can meet the harvest requirements.The comprehensive optimization results were verified by field experiments with different ridge shapes and different planting patterns.The results showed that the mean values of the missed excavation rate of different planting patterns of conventional straight ridges and extremely large"S"ridges were 2.23%and 2.69%,respectively,and the maximum values were 2.39%and 2.98%,respectively;the average damage rates were 3.38%and 4.14%,and the maximum values were 3.58%and 4.48%,which meet the industry standards of sugar beet harvester operation quality.The overall adaptability of the automatic row following system is good.This study can provide a reference for research on automatic row following harvesting systems of sugar beets and other subsoil crop harvesters.展开更多
利用支持向量机采用的结构风险最优化准则、预测能力强、鲁棒性好等优点,研究了最小二乘支持向量机(Least Squares Support Vector Machines,LS-SVM)回归算法在曳引机故障预测中的应用。提出了一种自动搜寻最优参数方法,对参数和进行寻...利用支持向量机采用的结构风险最优化准则、预测能力强、鲁棒性好等优点,研究了最小二乘支持向量机(Least Squares Support Vector Machines,LS-SVM)回归算法在曳引机故障预测中的应用。提出了一种自动搜寻最优参数方法,对参数和进行寻优,避免了人工选择的盲目性,提高了算法的效率。通过将LS-SVM和RBF神经网络进行对比实验,得出在相同训练样本条件下,LS-SVM可以取得比RBF更好的预测精度和预测速度,更加适合于现场实际应用。最后将LS-SVM模型用于曳引机振动信号的时域分量预测中,预测的平均相对误差小于5%,取得了较高的预测精度。展开更多
基金Supported by the Scientific Research Foundation for the Imported Talents(YKJ201014)~~
文摘In order to truly obtain the feature extraction of vibration signals under the strong background noise, the analysis and improvement of empirical mode decomposition (EMD) is carried on. After that, the improved EMD is applied to the feature extraction of vehicle vibration signals. First, the multi-autocorrelation method is adopted in each input signal,so the noise is reduced effectively. Then, EMD is used to deal with these signals,and the intrinsic mode functions (IMFs) are obtained. Finally, for obtaining the feature information of these signals, the Hilbert transformation and the spectrum analysis are performed in some IMFs. Theoretical analysis and ex- periment verify the effectiveness of the method, which are valuable reference for the same engineering problems.
基金supported by the National Natural Science Foundation of China(Grant No.52105263)the Key Laboratory of Intelligent Equipment and Robotics for Agriculture of Zhejiang Province(Grant No.2022ZJZD2201).
文摘To improve the automation level and operation quality of China's beet harvester and reduce the loss due to damaged and missed excavation,this study used a self-developed sugar beet combine harvester and field simulation experiment platform,based on the single-factor bench test of the automatic row following system in the early stage,taking hydraulic flow A,spring preload B,and forward speed C which have significant influence on performance indices as test factors,and taking the missed excavation rate,breakage rate and reaction time as performance indices,the orthogonal experimental study on the parameter optimization of the three-factor and three-level automatic row following system with the first-order interaction of various factors was carried out.The results of the orthogonal experiments were analyzed using range analysis and variance analysis.The results showed that there were differences in the influence degree,factor priority order and first-order interaction,and the optimal parameter combination on each performance index.A weighted comprehensive scoring method was used to optimize and analyze each index.The optimal parameter combination of the overall operating performance of the automatic row following system was A 2B 2C 1,that is,the hydraulic flow was 25 L/min,the forward speed was 0.8 m/s,and the spring preload was 198 N.Under this combination,the response time was 0.496 s,the missed excavation rate was 2.35%,the breakage rate was 3.65%,and the operation quality was relatively good,which can meet the harvest requirements.The comprehensive optimization results were verified by field experiments with different ridge shapes and different planting patterns.The results showed that the mean values of the missed excavation rate of different planting patterns of conventional straight ridges and extremely large"S"ridges were 2.23%and 2.69%,respectively,and the maximum values were 2.39%and 2.98%,respectively;the average damage rates were 3.38%and 4.14%,and the maximum values were 3.58%and 4.48%,which meet the industry standards of sugar beet harvester operation quality.The overall adaptability of the automatic row following system is good.This study can provide a reference for research on automatic row following harvesting systems of sugar beets and other subsoil crop harvesters.
文摘利用支持向量机采用的结构风险最优化准则、预测能力强、鲁棒性好等优点,研究了最小二乘支持向量机(Least Squares Support Vector Machines,LS-SVM)回归算法在曳引机故障预测中的应用。提出了一种自动搜寻最优参数方法,对参数和进行寻优,避免了人工选择的盲目性,提高了算法的效率。通过将LS-SVM和RBF神经网络进行对比实验,得出在相同训练样本条件下,LS-SVM可以取得比RBF更好的预测精度和预测速度,更加适合于现场实际应用。最后将LS-SVM模型用于曳引机振动信号的时域分量预测中,预测的平均相对误差小于5%,取得了较高的预测精度。